Just to jump off from this point and make a guess about your question...
given the large number of parameters relative to your sample size, you will
undoubtedly have some bootstrap samples that contain no observations for
some clusters, leading to one or more dummy variables and probably some of
your other variables droppping out of the estimation in certain samples. If
the ordering of the variables in the model affects which variables Stata may
drop in a given bootstrap sample (given a choice among several), then you
won't get replicable results with different orderings even after setting the
seed. Of course, this may not be very relevant given Stas' point below.
Michael Blasnik
[email protected]
----- Original Message -----
From: "Stas Kolenikov" <[email protected]>
To: <[email protected]>
Sent: Friday, July 30, 2004 1:53 PM
Subject: Re: st: Is there a fixed effect quantile regression in STATA?
> > My dependent variable is diff, and my independent variables are
> > diff3, trend and lmis_. I have 129 subjects identified by the
> > variable id. So I run two identical models with different sequencing
> > of the independent variables using the latest updated Stata 8, and
> > find that although the coefficients are consistent, the standard
> > errors are different (see below). I would be eager to send you my
> > dataset if you would like to have a closer look into the matter.
>
> I've never worked with -sqreg-, but unless the bootstrapping procedure
> inside it can take care of the clustering on id, the standard errors are
> pretty much useless (as the bootstrap is with the dependent data unless
> you make a Herculian effort to take the dependence into your resampling
> procedure). And I don't see -id- appearing anywhere in your -sqreg-
> command. Just curious.
>
> --- Stas Kolenikov
> -- Ph.D. student in Statistics at UNC-Chapel Hill
> - http://www.komkon.org/~tacik/ -- [email protected]
>
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
